iMac keeps crashing after Catalina upgrade

GWithers wrote:

Y'all resolve this issue?

I'm running a mid-2018 MBP 2.9 GHz 6-Core Intel Core i9, Radeon Pro Vega 20 4 GB and am having the same issue since upgrading to Catalina 10.15.3.

While I can't confirm that it only happens when I use my Caldigit T3 dock with one monitor connected via DisplayPort, it does seem more frequent than when I have them connected through two different Mac and Caldigit HDMI/Displayport adapters.

I spend my days editing documentaries with tons of footage but with proxies so while there is a Thunderbolt RAID attached, it's not doing crazy work...

I've been monitoring RAM and CPU usage and nothing seems too out of whack - my suspicion is that the GPU is getting overheated between the 2 monitors and Premiere and that's what's causing the issue.

Any thoughts?

There's a firmware update for the CalDigit TS3+ dock that seems to help some with this (and bumps power output to 87W so it doesn't trash your battery as badly).

It's less frequent, but it still happens. I'm currently up to nearly a week without a crash with 'Automatic graphics switching' turned off. YMMV. Quite an achievement for a machine so absurdly expensive (2019 MBP w/ Vega20) /s.

For me at least, I can confirm the crashing still happens without the dock in the chain. Also with the updated firmware, and also on the newest version of macOS.

This situation is really sad, and I've started the move away from Apple PC hardware.

For those getting the watchdogd timer panic, when you enter

kextstat | grep -v com.apple

to Terminal, is there anyone who gets nothing from the command, ie, you have no non-Apple kext?

My panics happen only when I have both Sophos and Little Snitch kexts installed, which proves nothing.

drdaz wrote:

It's less frequent, but it still happens. I'm currently up to nearly a week without a crash with 'Automatic graphics switching' turned off. YMMV. Quite an achievement for a machine so absurdly expensive (2019 MBP w/ Vega20) /s.

My mileage just varied. It crashes with those settings too.

vonlost wrote:

For those getting the watchdogd timer panic, when you enter

kextstat | grep -v com.apple

to Terminal, is there anyone who gets nothing from the command, ie, you have no non-Apple kext?

My panics happen only when I have both Sophos and Little Snitch kexts installed, which proves nothing.

It apparently proves that your computer doesn't crash without those 2 kexts loaded.

My output from that command is:

➜  Desktop kextstat | grep -v com.apple
Index Refs Address            Size       Wired      Name (Version) UUID <Linked Against>
  164    0 0xffffff7f847b0000 0x4000     0x4000     com.intel.driver.EnergyDriver (3.7.0) 35E739F9-BF6C-3024-A67C-750711B3FB64 <8 6 5 3>

EDIT: Just checked and that kext is part of Intel Power Gadget. It was crashing before I installed it.

I don't get anything when I run that command but I'm still getting the panics approx. every 10 days since 10.15.3 (and prior to that much more frequently despite a logic board, HDD and power supply replacement):

kextstat | grep -v com.apple

Index Refs Address            Size       Wired      Name (Version) UUID <Linked Against>

Going from Catalina to Mojave was the best macOS upgrade yet.

I had forgotten how stable my iMac Pro could be!

In all seriousness, with so many of us having Catalina problems which are fully fixed by rolling back to the previous OS, is there anything more we can do to raise visibility for Apple QA? When this reproduces even on virgin OS installs (no extra kexts or privileged software), surely there's more that Apple can do? Software that's running 100% in user mode shouldn't be able to panic the kernel, even if the developers tried to make it do so.

Always WindowServer/watchdogd:

panic(cpu 2 caller 0xffffff7f9ca3fad5): userspace watchdog timeout: no successful checkins from com.apple.WindowServer in 120 seconds

service: com.apple.logd, total successful checkins since load (69961 seconds ago): 6997, last successful checkin: 0 seconds ago

service: com.apple.WindowServer, total successful checkins since load (69931 seconds ago): 6981, last successful checkin: 120 seconds ago

Backtrace (CPU 2), Frame : Return Address

0xffffffa3ed3d3820 : 0xffffff801b93bb1b 

0xffffffa3ed3d3870 : 0xffffff801ba733e5 

0xffffffa3ed3d38b0 : 0xffffff801ba64e5e 

0xffffffa3ed3d3900 : 0xffffff801b8e2a40 

0xffffffa3ed3d3920 : 0xffffff801b93b207 

0xffffffa3ed3d3a20 : 0xffffff801b93b5eb 

0xffffffa3ed3d3a70 : 0xffffff801c0d2575 

0xffffffa3ed3d3ae0 : 0xffffff7f9ca3fad5 

0xffffffa3ed3d3af0 : 0xffffff7f9ca3f7e6 

0xffffffa3ed3d3b10 : 0xffffff801c066fcb 

0xffffffa3ed3d3b60 : 0xffffff801c070083 

0xffffffa3ed3d3ca0 : 0xffffff801ba22c22 

0xffffffa3ed3d3db0 : 0xffffff801b941998 

0xffffffa3ed3d3e10 : 0xffffff801b918625 

0xffffffa3ed3d3e70 : 0xffffff801b92f0d5 

0xffffffa3ed3d3f00 : 0xffffff801ba4b485 

0xffffffa3ed3d3fa0 : 0xffffff801b8e3226 

      Kernel Extensions in backtrace:

         com.apple.driver.watchdog(1.0)[EFAA17B9-A62C-30B7-9602-0BB5282EDDD6]@0xffffff7f9ca3e000->0xffffff7f9ca46fff

BSD process name corresponding to current thread: watchdogd

Ok, you have the "userspace watchdog timeout" instead of the "watchdogd" panic that I have, that I associate with non-Apple kexts. However, I recently switched my iMac from 5GHz wifi to Ethernet and haven't had a panic since. I also took 10.15.4 but others say it didn't make any difference. So I really don't know anything more.

I keep having " Kernel Panics " and from what I see posted in this list it looks like the major problem is :

1 - Catalina

2 - video driver

I have yet not tried going back to Mojave

I do not get the Panic Attacks that often but the point is they happen none the less, Apple know it and their is no fix ?!?

iMac (Retina 5K, 27-inch, 2019) 40 GB RAM

3.6 GHz 8-Core Intel Core i9

Radeon Pro 580X 8 GB

2 - 27" Thunderbolt Displays

I have been sending crash reports since upgrading to Catalina. This has been ongoing and relatively unchanged since the upgrade. I am sure others have also been sending in crash reports too. This leads me to the conclusion that either Apple does not view the reports or they cannot resolves the issues causing Catalina to hang on nearly ALL compatible Macs.

I did get some improvement when I read about some people using Norton Security, I contacted Symantec and they knew of the problem, said something about when the System goes into screen saver / scan when idle would sometimes cause the problem. they upgraded me to Norton 360 but then about 2/3 weeks later it happened again.

I guess my frustration is there is no " Sure Fire Way " to repeat it, at least on my system.

It sounds like calling Apple / AppleCare is not going to do me any good

I have just submitted my 100th crash report since I stared keeping a copy as a text file as well as sending to Apple.

Twice already this morning on my MBP :/

1:

panic(cpu 0 caller 0xffffff80096a295c): Sleep transition timed out after 180 seconds while calling power state change callbacks. Suspected bundle: com.apple.iokit.IOSkywalkFamily. Thread 0x12e42.
Failure code:: 0x00000008 00000014

2:

panic(cpu 4 caller 0xffffff7f91cb2ad5): userspace watchdog timeout: no successful checkins from com.apple.logd in 120 seconds
service returned not alive with context : unresponsive dispatch queue(s): com.apple.firehose.io-wl 

Radeon Pro 560 in my MBP

Part of the stability of Apple products compared to WinPC is that there is a known set of hardware, so easy for Apple to test against every commercially available combination, rather than the trillions of combos of motherboard, cpu, gpu, SSD, RAM etc etc

At least on my Windows machine I can go direct to the vendor and search for an updated driver patch, or simply roll back that specific part to a known working state.

Apple products "just work", until they don't - at which point you are utterly borked.

Just curious Sir, are you running multiple monitors ?

I ask because since I have been having these " Panics " it usually happens when my system does something related with screensaver or waking . I have multiple monitors.

but the strange part is there is not a particular app running in the background to add as a 3rd party ( if that makes sense )

In other words it is like there is multiple things going on that cause it,

That is why I asked about if anyone has disabled SIP, I know the system is at times incredibly paranoid.

Would love to get more information from you,

Thanks

They knew about this when I called them. So you know Apple is aware of it as well

I was having Panic Attacks as I called them 5 plus times a week, but I was chalking it up to running the Catalina Beta as well.

I used one of my SSD to create another external clean system, Catalina 10.15.4 this time no Symantec and a bunch of other odds and ends I normally use.

I have been using this system, which I am replying on BTW to just stream Netflix and stuff while surfing the web and guess what .... Panic Attack as random as could be

I am seriously considering disabling SIP, I know it is not a wise but maybe the System is just being too paranoid.

Would like to find a reason why Mohave works and Catalina is freaking out

A straight answer would be golden or even " Hey don't do this stuff at the same time because there is an issue " but I am betting that this will not be solved until the next Major OS release